Yep,
This was 100% my fault here sorry. Turns out we add the repo as part of
our bootstrap process and I just missed the line.
Sorry for the confusion!
Cheers,
James
On Tue, 24 Jan 2017, at 06:24 PM, Yedidyah Bar David wrote:
On Tue, Jan 24, 2017 at 7:00 AM, James <james(a)jloh.co> wrote:
> Hi guys,
>
> I'm trying to servers to a new oVirt 4.0 cluster. I've upgraded the
> engine (not hosted, standalone) to version 4.0.6.3 and run engine-setup
> and everything seemed to go fine.
>
> When reinstalling oVirt on a host after removing it from the 3.6 compat
> cluster into a 4.0 compat cluster causes issues when reinstalling and
> setting up the management network for some reason. I haven't been able
> to track this issue down fully yet.
>
> Whats weird is that even when a fresh new host is installed and added to
> the grid for some reason oVirt installs the repos for oVirt 3.6 on the
AFAIK nothing in oVirt installs repos. It's up to the admin to add the
correct repos, usually by installing a suitable ovirt-release*.rpm
package.
Please check all of /etc/yum.repos.d before you (re)install such a host.
> host. I'm sure this is done by oVirt itself since they aren't there
> before I install the host and we don't manage these outside of that. It
Can you please verify that somehow? E.g. 'yum search vdsm', 'yum info
vdsm'?
Also, which OS are you using?
> also just stalls at "Activating" the host and I can't do anything
with
> them after that.
>
> Shouldn't it be installing version 4.0 for this cluster? I've triple
> checked its set to version 4.0 compatibility and the data center its in
> is also set to this level.
>
> Unsure what logs I can provide but any help would be greatly
> appreciated!
Generally speaking, you should check:
On engine machine - /var/log/ovirt-engine/* ,
/var/log/ovirt-engine/host-deploy/*
On host - /var/log/vdsm/* , during install also /tmp/*
But first check your repos.
Best,
--
Didi